MCPcopy Create free account

hub / github.com/Tencent/TurboTransformers / functions

Functions392 in github.com/Tencent/TurboTransformers

⨍ Functions 392◇ Types & classes 103
↓ 1 callersFunctionGetPoolType
turbo_transformers/layers/kernels/seq_pool.cpp:161
↓ 1 callersFunctionInitMemcpyFuncs
turbo_transformers/core/memory.cpp:34
↓ 1 callersMethodIsExist
turbo_transformers/loaders/npz_load.h:40
↓ 1 callersFunctionLoadEmbedding
example/cpp/bert_model.cpp:33
↓ 1 callersFunctionLoadPooler
example/cpp/bert_model.cpp:43
↓ 1 callersFunctionMatmulBenchmarkGeneralHelper
benchmark/turbo_transformers/layers/kernels/matmul_benchmark.cpp:84
↓ 1 callersFunctionNewDLPackTensor
turbo_transformers/core/tensor.cpp:41
↓ 1 callersFunctionSequence
turbo_transformers/layers/kernels/common.cpp:38
↓ 1 callersMethodSequencePool
turbo_transformers/layers/sequence_pool.h:25
↓ 1 callersFunctionSoftmaxMask
attr_mask's shape could be (batch, from_len, to_len), (batch, 1, to_len) or nullptr is2D is used to distinguish the two scenarios.
turbo_transformers/layers/kernels/softmax.cpp:28
↓ 1 callersMethodToDLPack
turbo_transformers/core/tensor.h:125
↓ 1 callersFunctionTransform
TODO(jiaruifang): this function should better pass a function in. how can we pass a lambda function as __device__ to cuda?
turbo_transformers/layers/kernels/common.cpp:76
↓ 1 callersFunctionTransposeForScoreImpl
tranpose(2,3)
turbo_transformers/layers/kernels/transpose.cpp:58
↓ 1 callersMethodWait
turbo_transformers/core/cuda_device_context.cpp:30
↓ 1 callersMethod_init_cache
(self, memory_bank)
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:608
↓ 1 callersFunctionbenchmark_torch
(model_name: str, seq_len: int, batch_size: int, n: int, enable_random: bool, max_seq_len:
benchmark/torch_benchmark_helper.py:17
↓ 1 callersFunctionbenchmark_torch_jit
(model_name: str, seq_len: int, batch_size: int, n: int, enable_random: bool, max_seq_
benchmark/jit_benchmark_helper.py:21
↓ 1 callersFunctionbenchmark_turbo_transformers
(model_name: str, seq_len: int, batch_size: int, n: int, enable_random: bool,
benchmark/turbo_benchmark_helper.py:17
↓ 1 callersFunctionbuild_dic
(num_layers)
tools/convert_huggingface_bert_tf_to_npz.py:23
↓ 1 callersMethodcheck_torch_and_turbo
(self, use_cuda)
turbo_transformers/python/tests/albert_layer_test.py:58
↓ 1 callersMethodcompute_major
turbo_transformers/core/cuda_device_context.cpp:40
↓ 1 callersFunctioncreate_shape_test
(batch_size: int, seq_length: int)
turbo_transformers/python/tests/bert_output_test.py:27
↓ 1 callersFunctioncreate_test
(batch_size, seq_length)
turbo_transformers/python/tests/bert_attention_test.py:28
↓ 1 callersFunctioncreate_test
(batch_size, seq_length)
turbo_transformers/python/tests/albert_attention_test.py:28
↓ 1 callersFunctioncreate_test
(batch_size, seq_length)
turbo_transformers/python/tests/bert_intermediate_test.py:27
↓ 1 callersFunctioncreate_test
(batch_size, seq_length)
turbo_transformers/python/tests/albert_layer_test.py:27
↓ 1 callersFunctioncreate_test
(batch_size, seq_length)
turbo_transformers/python/tests/bert_layer_test.py:29
↓ 1 callersFunctioncreate_test
(batch_size, seq_length)
turbo_transformers/python/tests/albert_model_test.py:27
↓ 1 callersFunctioncreate_test
(batch_size, src_length, T, with_quantize_dynamic=False)
turbo_transformers/python/tests/decoder_transformer_decoder_layer_test.py:28
↓ 1 callersFunctioncreate_test
(batch_size)
turbo_transformers/python/tests/bert_pooler_test.py:27
↓ 1 callersFunctioncreate_test
(batch_size, input_len)
turbo_transformers/python/tests/positionwise_ffn_test.py:29
↓ 1 callersFunctioncreate_test_bert_emb
(batch_size: int, seq_length: int)
turbo_transformers/python/tests/bert_embedding_test.py:25
↓ 1 callersFunctioncreate_test_bert_emb
(batch_size: int, seq_length: int)
turbo_transformers/python/tests/albert_embedding_test.py:38
↓ 1 callersFunctioncreate_test_seq_pool
(batch_size: int, seq_length: int, pool_type: str)
turbo_transformers/python/tests/sequence_pool_test.py:34
↓ 1 callersFunctioncuda_alloc
turbo_transformers/core/allocator.cpp:40
↓ 1 callersFunctioncuda_free
turbo_transformers/core/allocator.cpp:53
↓ 1 callersMethodfrom_torch
Args: model : a PyTorch GPT2 Model device : cpu or GPU backend : a string to indicates kernel providers
turbo_transformers/python/turbo_transformers/layers/modeling_gpt2.py:75
↓ 1 callersFunctionget_extended_attention_mask
Makes broadcastable attention mask and causal mask so that future and maked tokens are ignored. Arguments: attention_mask: torch.Tensor wi
turbo_transformers/python/turbo_transformers/layers/utils.py:86
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/bert_attention_test.py:30
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/albert_attention_test.py:30
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/bert_embedding_test.py:27
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/decoder_multi_headed_attention_test.py:38
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/bert_model_test.py:27
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/bert_intermediate_test.py:29
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/albert_layer_test.py:29
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/albert_embedding_test.py:40
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/roberta_model_test.py:27
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/bert_layer_test.py:31
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/albert_model_test.py:29
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/decoder_transformer_decoder_layer_test.py:30
↓ 1 callersMethodinit_data
(self, use_cuda: bool)
turbo_transformers/python/tests/bert_pooler_test.py:29
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/positionwise_ffn_test.py:31
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/bert_output_test.py:29
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/bert_encoder_test.py:27
↓ 1 callersMethodinit_data
(self, use_cuda)
turbo_transformers/python/tests/gpt2_model_test.py:27
↓ 1 callersFunctionmain
()
tools/convert_huggingface_bert_pytorch_to_npz.py:23
↓ 1 callersFunctionmain
()
tools/convert_huggingface_bert_tf_to_npz.py:63
↓ 1 callersFunctionmain
()
benchmark/benchmark.py:42
↓ 1 callersFunctionmain
()
benchmark/benchmark_result_to_csv.py:20
↓ 1 callersFunctionpooling
(input, pool_type)
turbo_transformers/python/tests/sequence_pool_test.py:20
↓ 1 callersMethodprint_results
turbo_transformers/core/profiler.cpp:134
↓ 1 callersMethodset_name
turbo_transformers/core/profiler.cpp:112
↓ 1 callersFunctiontest
(loadtype: LoadType, use_cuda: bool)
example/python/gpt2_example.py:27
↓ 1 callersFunctiontest
(use_cuda)
example/python/roberta_example.py:23
↓ 1 callersFunctiontest
(loadtype: LoadType, use_cuda: bool)
example/python/albert_example.py:27
↓ 1 callersFunctiontest
(loadtype: LoadType, use_cuda: bool)
example/python/bert_example.py:26
↓ 1 callersFunctiontrans_layer_name_tf2turbo
(dic, name)
tools/convert_huggingface_bert_tf_to_npz.py:59
FunctionActivationTestHelper
turbo_transformers/layers/kernels/activation_test.cpp:30
FunctionAddBiasAct
turbo_transformers/layers/kernels/activation.cpp:91
FunctionAddBiasLayerNorm
turbo_transformers/layers/kernels/layer_norm.cpp:99
FunctionAddBiasTestHelper
turbo_transformers/layers/kernels/utils_test.cpp:73
FunctionAddInputBiasTestHelper
turbo_transformers/layers/kernels/utils_test.cpp:102
MethodAlbertLayer
turbo_transformers/layers/albert_layer.h:24
MethodAllocator
* APIs of Allocator *********/
turbo_transformers/core/allocator.cpp:202
MethodBERTEmbedding
turbo_transformers/layers/bert_embedding.h:25
MethodBERTLayer
example/cpp/bert_model.cpp:52
MethodBadAlloc
turbo_transformers/core/allocator.cpp:32
MethodBertAttention
turbo_transformers/layers/bert_attention.h:27
MethodBertIntermediate
turbo_transformers/layers/bert_intermediate.h:24
MethodBertOutput
turbo_transformers/layers/bert_output.h:24
MethodBertPooler
turbo_transformers/layers/bert_pooler.h:24
MethodBestFitAllocatorImpl
turbo_transformers/core/allocator.cpp:127
MethodCPUTimer
benchmark/turbo_transformers/layers/kernels/benchmark_help.h:56
MethodCUDADeviceContext
turbo_transformers/core/cuda_device_context.cpp:23
MethodCachingAllocatorImpl
turbo_transformers/core/allocator.cpp:137
FunctionCallBackFunction
example/cpp/bert_model_example.cpp:45
FunctionCallBackFunction
example/cpp/bert_model_test.cpp:112
FunctionCheckResultOfCPU
turbo_transformers/layers/kernels/common.h:135
FunctionCheckResultOfCPUAndGPU
turbo_transformers/layers/kernels/common.h:113
FunctionConcat
turbo_transformers/layers/kernels/utils.cpp:33
FunctionConcatTestHelper
turbo_transformers/layers/kernels/gpu_utils_test.cpp:30
FunctionCreateAndFillRandomForCPUGPUTensors
turbo_transformers/layers/kernels/common.h:103
FunctionCreateTensor
turbo_transformers/layers/kernels/common.h:70
FunctionCreateTensorAndFillConstant
turbo_transformers/layers/kernels/common.h:79
FunctionCreateTensorAndFillRandom
turbo_transformers/layers/kernels/common.h:89
FunctionDLManagedTensorDeletor
turbo_transformers/core/tensor.cpp:23
FunctionDLPack_Capsule_Destructor
turbo_transformers/python/pybind.cpp:41
FunctionDisableGperf
turbo_transformers/core/profiler.cpp:163
MethodElapseSecond
benchmark/turbo_transformers/layers/kernels/benchmark_help.h:60
FunctionEnableGperf
turbo_transformers/core/profiler.cpp:152
← previousnext →101–200 of 392, ranked by callers